31-01-2013, 09:42 AM
Wheelchair Motion Control Guide Using Eye Gaze and Blinks Based on PointBug Algorithm
1Wheelchair Motion Control.pdf (Size: 452.26 KB / Downloads: 147)
Abstract
In this paper, we propose new method beside the
classic method, to control the motorized wheelchair using EOG
signals. The new method allows the user to look around freely
while the wheelchair navigates automatically to the desired
goal point. Only EOG signals are used to control the
wheelchair; eye gazing and blinking. The user can still choose
to control the wheelchair using the classic manual method in
case the environment and obstacles structure does not help
with the auto navigation method. In the new auto navigation
method the microcontroller can know the goal point direction
and distance by calculating the gaze angle that the user is
gazing at. PoingBug algorithm is used to navigate the
wheelchair in Auto controlling method. Simulated results are
similar to Tangent Bug algorithm results, but experimental
tests are slightly improved in some cases where the
surroundings have sharp edges.
INTRODUCTION
It is estimated by New Freedom Initiative Act that,
disabled and elderly people who need a wheelchair are to be
over 100 million worldwide. A lot of these disabled cannot
use the traditional electrical wheelchairs, which are
controlled by joystick, due to their limb movement
restrictions. Hence, many alternative control methods have
been developed to allow those disabled users to live more
independently, e.g. voice recognition and guidance [1] and
[2], vision based head gesture control [3]-[7], EMG signal
based control [8] and [9], and EOG eye tracking control [10]
and [11]. Even though, EOG eye tracking control offers a
more natural mode to guide the wheelchair, it is omitted
because users are normally not allowed to look around the
surrounding environment during motion. Most researches
tried to overcome this problem by using EMG along with
EOG signals to control the wheelchair [12] and [13].
Paper Scope and Contributions
In this paper, we present EOG signals based control
method of the wheelchair. A hands-free control system that
uses only EOG signals for control of an electric powered
wheelchair. New automatic controlling method is introduced
besides the common manual method. The user can look
around the surrounding environment freely during the
navigation process in automatic mode.
RELATED WORK
A. Wheelchair mation control methods
A powered wheelchair is driven by electric motors,
instead of the user’s arms. It is normally controlled with a
joystick. However, other input devices such as voice, EOG
and EMG are used if the user has restricted limbs
movements.
This work has proposed an eye tracking based on EOG
signals to control the wheelchair. Wheelchair motion control
using eye tracking has been applied in many research studies.
For example, [12] and [13] used EMG to control the
direction and EOG to control the speed of the wheelchair.
While other researches has focused on Human Computer
Interfaces (HCI) based on EOG signals, where the user will
be allowed to perform other simple tasks beside wheelchair
motion control [14]. Although, only few focused on using
EOG signals solely to control the wheelchair [15] and [16].
On the other side, current studies describe the design
approaches for gaze control, where the user controls the
wheelchair by gazing directly at the physical target point
[17] and [18]. However, the eye tracking method used was
camera based which is more expensive, complex, and the
camera must be mounted to the head of the user.
PROPOSED NEW TECHNIQUES
Previous researches use classical manual method to
control the wheelchair motion using EOG signals. In manual
method, the user look up to move forward, right to turn right,
left to turn left, and down to stop the wheelchair. This
method exhausts the user due to the concentration needed
during the navigation process. In addition to that, the user is
not allowed to look around the surrounding environment
while navigating the wheelchair to the desired destination.
This research proposes to get rid of this drawback by
introducing new controlling method, which takes advantage
of the gaze angle detection. In the new method, the user only
needs to look at the desired destination, and then blink to
give the signal to the controlling unit to start navigation.
After that, the wheelchair will calculate the desired goal
position and distance from the measured gaze angle. Finally,
the wheelchair will move toward the destination in straight
line and go around obstacles when detected by sensors (see
Fig. 3).
Navigation control method
In this research, the user can choose between two
methods to control the wheelchair. The first method is the
common manual method, where the user look up to go
straight, look right to turn right, look left to turn left, and
look down to stop. The user cannot look around the
surroundings while controlling the wheelchair using this
method.
However, the other method provides the freedom for the
user to look around freely during the navigation process. The
user only needs to blink after gazing at the desired
destination. After that, the wheelchair will automatically
navigate to the target position. The gaze angle can be
measured by the EOG measurement system. Intentional
blinks are used as input signals. They are commands to
switch between the controlling methods and to execute
navigation process.
Conclusions
This paper presents EOG signals measurement system in
order to control the wheelchair motion using only eye
gazing and blinking. The main objectives of this project is
to help disabled, who can only move their eyes to easily
control the wheelchair, and yet look around the surrounding
environments while navigation process is done
automatically. Particular hardware has been developed to
capture users’ biopotentials.
Gaze angles between 0° and 40° with increment of 10°
are easily measured and by measurement system and
detected by controlling unit as inputs. Intentional blinks are
also easily detected and used as input commands in the
controlling method.
PointBug algorithm is implemented and used to control
the wheelchair in auto navigation method. The simulation
shows that the user can enjoy looking around while the auto
controlling method navigate to the goal point with avoiding
obstacles, however there are cases where manual controlling
method is more efficient to use such as in map C case.